Pre-CVPR Seminar 2022

Description

The purpose of this seminar is to bring together people across the country with interest in computer vision and pattern recognition. The seminar features presentations and discussions around papers accepted at the 2022 Computer Vision and Pattern Recognition Conference or other computer vision related presentations.

Specific organisers 

• Dimitrios Papadopoulos dimp@dtu.dk
• Serge Belongie s.belongie@di.ku.dk
• Negar Heidari negar.heidari@ece.au.dk
  
• Sagie Benaim sabe@di.ku.dk
• Frederik Rahbæk Warburg frwa@dtu.dk

Tentative schedule

12:00 – 13:00 Arrivals and lunch

13:00 – 13:10
Welcome
Serge Belongie & Dimitrios Papadopoulos 

13:10 – 14:15  Session I

1. Tuukka Ruotsalo (UCPH, DK)
   Brain-Supervised Image Editing
2. Rene Haas (UCPH, DK)
   Tensor-based Emotion Editing in the StyleGAN Latent Space
3. Tong Yu (NTNU, NO)
   Paramixer: Parameterizing Mixing Links in Sparse Factors Works Better than Dot-Product Self-Attention
4. Sagie Benaim (UCPH, DK)
   Text2Mesh: Text-Driven Neural Stylization for Meshes
5. Manxi Lin (DTU, DK)
   diffConv: Analyzing Irregular Point Clouds with an Irregular View

12:00 – 13:00

Arrivals and lunch

13:00 – 13:10

Welcome: Serge Belongie & Dimitrios Papadopoulos

13:10 – 14:15

Session I: (10+5 min presentations)

1. Tuukka Ruotsalo (UCPH, DK)
Brain-Supervised Image Editing

2. Rene Haas (UCPH, DK)
Tensor-based Emotion Editing in the StyleGAN Latent Space

3. Tong Yu (NTNU, NO)
Paramixer: Parameterizing Mixing Links in Sparse Factors Works Better than Dot-Product Self-Attention

4.Sagie Benaim (UCPH, DK)
Text2Mesh: Text-Driven Neural Stylization for Meshes

5. Manxi Lin (DTU, DK)
diffConv: Analyzing Irregular Point Clouds with an Irregular View

14:15 – 14:45

Keynote: Professor Angela Dai, TU Munich. 

14:45 – 15:00

Break

15:00 – 16:05 

Session II:

1. Neelu Madan (AAU, DK)
Self-Supervised Predictive Convolutional Attentive Block for Anomaly Detection

2. Erik Gartner (Lund, SE)
Trajectory Optimization for Physics-Based Reconstruction of 3d Human Pose from Monocular Video

3. Rasmus Haugaard (SDU, DK)
SurfEmb: Dense and Continuous Correspondence Distributions for Object Pose Estimation with Learnt Surface Embeddings

4. Dindging Cai (Tampere, FI)
OVE6D: Object Viewpoint Encoding For Depth-based 6D Object Pose Estimation

5. Karsten Roth (Tübingen, DE)
Non-isotropy Regularization for Proxy-based Deep Metric Learning

16:05 – 16:20

Break

16:20 – 17:30

Session III:

1. Dim Papadopoulos (DTU, DK)
Learning Program Representations for Food Images and Cooking Recipes

2. Karsten Roth (Tübingen, DE)
Integrating Language Guidance into Vision-based Deep Metric Learning

3. Almut Sophia Koepke (Tübingen, DE)
Audio-visual Generalised Zero-shot Learning with Cross-modal Attention and Language

4. Jae Myung Kim (Tübingen, DE)
Large Loss Matters in Weakly Supervised Multi-Label Classification

5. Massimiliano Mancini (Tübingen, DE)
KG-SP: Knowledge Guided Simple Primitives for Open World Compositional Zero-Shot Learning